Thursday, August 12, 2010

Thoughts on TIPS and gold as inflation hedges



TIPS (Treasury Inflation Protected Securities, the original acronym) are complex bonds that are difficult to evaluate. I first recommended TIPS in October 2008 ("TIPS are a steal"), and pounded the table to buy them for the next several months. I have been consistently a fan of TIPS (though more recently only for as a substitute for cash) ever since. If you take TIP as a proxy for the TIPS market, TIPS have enjoyed price appreciation of some 18% since their low in November '08, plus interest (in the form of their real coupon) of about 5%, for a total return of roughly 24%. That's not as good as equities, but then TIPS are default-free bonds that pay guaranteed interest while equities are full of risk.

As the chart above shows, the real yield on TIPS is now as low as it's ever been, which means that TIPS prices are at or near their all-time highs (the market price of TIPS varies inversely to their real yield). As my chart also suggests, TIPS are in "expensive" territory based on their real yield. Is now the time to sell?

If you are trader holding TIPS and not particularly concerned about the future of inflation or downside risk to the economy, then now is an excellent time to sell TIPS. Real yields might fall a bit further, but I think they are reaching their limit. Real yields are low because nominal yields are low, and both are low because the market is terribly concerned about the risk of a significant slowdown in the U.S. economy. Low Treasury and TIPS yields are a direct reflection of a deep pessimism that pervades both the stock and the bond markets. I don't share the market's pessimism, so I'm willing to think that the decline in yields has just about run its course.

If you are a long-term investor holding TIPS, then you should understand that TIPS are quite unlikely to enjoy any price appreciation from here. TIPS returns will be driven entirely by the change in the CPI going forward, and the risk of some price depreciation in the future is not negligible. As the next chart shows, TIPS are priced (relative to Treasuries) to the assumption that the CPI is going to rise 1.8% per year on average for the foreseeable future. (That's the breakeven or expected inflation rate.) Thus, the market is saying that a TIPS holder is likely to earn 1.8% per year from inflation, plus a 1% coupon, for a total return of 2.8% per year, minus any price depreciation that happens to occur.

That's not a particularly exciting prospect. TIPS are going to deliver exciting returns ONLY if inflation turns out to be significantly higher than the market expects, and they could deliver disappointing returns if inflation doesn't change much, if real yields rise because TIPS fall out of favor, and/or the economy picks up and Treasury yields in general rise.

So TIPS are now a pure bet on inflation. Inflation has to pick up meaningfully for an investment in TIPS to do better than Treasury bonds. If it doesn't, you're going to either make a pittance or you're going to lose some money.

I could say the same thing about gold, however, but in spades. Consider that gold has soared 370% since its early 2001 low (next chart). If gold is up because central banks all over the world are in accommodation mode, then gold is anticipating lots of inflation down the road. If you buy gold today, you NEED lots of inflation to show up, and you NEED central banks to remain super-accommodative, otherwise gold could collapse. If gold is up because of geopolitical risks or other types of raw fear, then you NEED things to never get better, otherwise gold could collapse.


TIPS and gold are two classic inflation hedges, but with very different characteristics. Gold is like a highly leveraged play on inflation, but TIPS have limited downside risk. Gold might drop by 60% or more. But TIPS have no default risk and, more importantly, the real yield on TIPS almost certainly has an upper limit (unlike the yield on nominal bonds, which can theoretically rise without limit). I say that because the real yield on TIPS is a U.S. government-guaranteed real yield—a risk-free real yield. At the end of the day, the only yield or return on an investment that really counts is its real (after-inflation) yield. If you could buy TIPS with a real yield of, say, 4% or more, would there be any investment in the world that could match that on a risk/reward basis? I seriously doubt it. If real yields on TIPS rise, at some point they will become absolutely compelling investments relative to the universe of other investments, and my guess is that the upper limit is around 4%.

If real yields were to rise from 1% to 4%, this would equate to a drop in price for TIP of 20-25%; that would be offset by their 1% real coupon plus whatever inflation happens to be. Thus, the downside risk to TIPS in a worst-case scenario (barring a U.S. government default, in which case all bets are off) would be a mark-to-market loss of 20-25% that could be reduced if one were willing to hold TIPS to maturity. (TIPS traded briefly above the 4% yield level in 2001, but that was when the TIPS market was still in its infancy, they were not well understood, and equities were all the rage because people thought the economy would grow 4-5% per year forever.)

To summarize: TIPS make sense only for long-term investors who are concerned about the risk of rising inflation and are willing to suffer some losses if that doesn't happen, and TIPS are a much less risky hedge against rising inflation than gold.

Full disclosure: I am still long TIP and TIPS at the time of this writing.

9 comments:

Benjamin said...
This comment has been removed by the author.
Benjamin said...

Excellent commentary, once again magnifying every ambivalent impulse I have.

My own guess is we see little no inflation for years in the United States, and perhaps mild deflation (I hope only mild).

Remember, ever since the 1970s the code for central bankers requires stout stands against inflation.

Public institutions do not adapt well, if at all, to changing circumstances, unlike private companies, which are forced to change continually.

The Fed is still fighting inflation--the last war.

Central Bankers in Japan still preach against inflation! They have wrecked their nation's economy, and still they think the goal is mild deflation. The yen has appreciated against gold for 20 years--and who benefitted? No one in Japan.

Gold is a whack-job investment, and good luck with it. As Grannis says, it may collapse, or maybe go higher on Chinese and Indian demand or the gold-nut crowd fever.

One thing we can be sure of: The Federal Reserve Board will continue with its pettifogging and sermonettes about inflation, even as we sink into deflation. That is what central bankers do.

Maybe you want to stuff cash under your mattress.

DaleW said...

Having re-read Peter Bernstein's The Power of Gold last month, I would say gold is a hedge against any kind of monetary instability, be it inflation or deflation. People hoard gold when they are fearful, which is a pattern of behavior that goes back to at least the ancient Near East cultures, based on my reading of Bernstein.

I recently priced five commodities in gold terms and all indicated gold is very richly priced, as you say. The first was US homes. In the post-Bretton Woods era, median home prices have averaged 307 ounces of gold. Right before the US closed the gold window in 1971, the median home price was 620 ounces. This fell to a low of 89 ounces in Q1 1980, on the cusp of one of the worst quarters for GDP growth since the Depression and amidst rising Cold War fears and double-digit inflation. In between, a huge rally in real estate prices took place from mid-1973 through mid-1976, but that faded until the early 1980s, after which residential real estate started a long march to a peak price of 586 ounces of gold in July, 2001. Since the 2001 peak, home prices have dropped by nearly 75% in gold terms, to 144 ounces, which we haven't seen in Q2 1981, in the midst of a brutal yet cleansing recession.

Going further back, I priced four other commodities in gold terms. These are freely purchased commodities, available widely, with a nearly constant quality, produced over the last 75 years by a consolidated group of producers, whose content mirrors a pretty basket of goods and services in the US and global economies.

The four commodities are Kellogg's Corn Flakes, Coca-Cola, Oreos, and a Hershey bar. Since these are advertised items for which the price and content (weight) are available for historians to study, I was able to price this on a per ounce basis to arrive at a price index over time. Two famous examples of this approach are the price of a quality men's suit in gold terms (since Roman times, a quality men's suit has cost roughly one ounce of gold, according to Stifel's Barry Bannister) and the Herengracht (Patrician's, Lord's, or gentleman's canal) real estate index (this refers to a canal in Amsterdam on which homes have always been owned by the upper middle class or higher, which takes hedonic adjustments out of the picture for long-term price inflation calculations, which is a seriously complicating factor in the calculation of any inflation rate).

Prices for the four goods above have increased by 4.7% annually since 1970, which equates to 0.8% pricing power in real terms if CPI is to be believed as a useful yardstick for inflation. I think CPI has been massaged over time and hedonic adjustments along with index composition muddle the picture. I would guess CPI understates inflation somewhat and that real pricing power for these commodities has been lower over this time, yet still positive.

Nevertheless, we can see by a common yardstick the price of these goods has gone down in gold terms (the proportion of an ounce of gold needed to buy an ounce of these products) by 26% since 1972. In addition, we are nearing the bottom of the 38-year post-Bretton Woods cycle of gold to an ounce of consumer packaged goods. That means gold is expensive in terms of consumer packaged goods and is perhaps overvalued according to this constant purchasing power parity index.

Setting purchasing power parity based on the 1972-2010 average of an ounce of gold to an ounce of consumer packaged goods, gold would be priced at $597.75 per ounce today, 50% below its last trade at $1,193 (also equal to gold's 9-10 year moving average).

This is another way of coming at Scott's conclusion. These goods either need to inflate massively to get gold back to its historical purchasing power or gold needs to come down by about 50%.

John said...

Inflation yes? But I would be more worried about things to come after that. Check this guy out, he called the crash back in 2008 and has been spot on with this for the last few years.

http://www.youtube.com/watch?v=gn6kS4l2yFM

Scott Grannis said...

Dale: Thanks for the extra analysis which supports my conclusions. One suggestion: use the 1960-65 period as your starting point instead of 1970. Monetary policy started to go off track right around 1965.

John said...

Dale,

Nice work. Thank you for your effort...good reading.

Guys,

The 'John' above is not me. No problem, just sayin'.

Mississippi Snopes said...

Scott, I am sorry to trouble you with a naive question, but could you direct me to a post where you explain the difference between "TIP" and "TIPS"? I'm considering moving a considerable percentage of my retirement assets to TIPS and want to make sure I understand this analysis completely. Thanks very much

Scott Grannis said...

M Snopes: "TIP" is the symbol for an ETF that invests only in TIPS, which are inflation-indexed T-bonds. TIP is a good way for the average investor to gain exposure to the TIPS market. There are a few other TIPS funds that would be equally as attractive as TIP if your aim is to gain a diversified exposure to the possibility that inflation rises by more than the market expects. You should do a search on my blog for "TIPS" and you will find numerous posts discussing their relative merits.

Mississippi Snopes said...

Thanks, Scott. I believe I've read most of those earlier posts, but just had one of those moments where I couldn't figure out what "TIP" means.

I'm interested in TIPS as insurance against loss of capital, not investing, so the TIP would not be as well suited to my needs.

Thanks very much for this page, which is by far my favorite financial site.